Enhanced inhibition of murine tumor and human breast tumor

Enhanced inhibition of murine tumor and humanbreast tumor xenografts using targeted deliveryof an antibody-endostatin fusion protein

Hyun-Mi Cho,1,3 Joseph D. Rosenblatt,1

Young-Sook Kang,4 M. Luisa Iruela-Arispe,5

Sherie L. Morrison,6 Manuel L. Penichet,6

Young-Guen Kwon,3 Tae-Woong Kim,3

Keith A. Webster,2 Hovav Nechustan,1

and Seung-Uon Shin1

1Department of Medicine, Hematology-Oncology, University ofMiami School of Medicine and Sylvester Comprehensive CancerCenter, and 2Department of Molecular and Cellular Pharmacology,University of Miami School of Medicine, Miami, Florida;3Department of Biochemistry, College of Natural Sciences,Kangwon University, Kangwon-do, Korea; 4College of Pharmacy,Sookmyung Women’s University, Seoul, Korea; and Departmentsof 5Molecular, Cell, and Developmental Biology and6Microbiology, Immunology, and Molecular Genetics, Universityof California at Los Angeles, Los Angeles, California

AbstractEndostatin can inhibit angiogenesis and tumor growth inmice. A potential limitation of endostatin as an anti-tumor agent in humans is the short serum half-life of theprotein that may decrease effective concentration at thesite of tumor and necessitate frequent dosing. In aneffort to improve antitumor activity, endostatin wasfused to an antibody specific for the tumor-selectiveHER2 antigen to create an antibody-endostatin fusionprotein (anti-HER2 IgG3-endostatin). Normal endostatinrapidly cleared from serum in mice (T1/2

2, = 0.6–3.8hours), whereas anti-HER2 IgG3-endostatin had a pro-longed half-life (90% intact; T1/2

2, 40.2–44.0 hours).Antigen-specific targeting of anti-HER2 IgG3-endostatinwas evaluated in BALB/c mice implanted with CT26tumors or CT26 tumors engineered to express the HER2antigen (CT26-HER2). Radio-iodinated anti-HER2 IgG3-endostatin preferentially localized to CT26-HER2 tumors

relative to CT26 tumors. Administration of anti-HER2IgG3-endostatin to mice showed preferential inhibition ofCT26-HER2 tumor growth compared with CT26. Anti-HER2 IgG3-endostatin also markedly inhibited thegrowth of human breast cancer SK-BR-3 xenografts insevere combined immunodeficient mice. Anti-HER2IgG3-endostatin inhibited tumor growth significantlymore effectively than endostatin, anti-HER2 IgG3 anti-body, or the combination of antibody and endostatin.CT26-HER2 tumors treated with the endostatin fusionprotein had decreased blood vessel density and branch-ing compared with untreated CT26-HER2 or CT26treated with the fusion protein. The enhanced effective-ness of anti-HER2 IgG3-endostatin may be due to alonger half-life, improved serum stability, and selectivetargeting of endostatin to tumors, resulting in decreasedangiogenesis. Linking of an antiangiogenic protein, suchas endostatin, to a targeting antibody represents apromising and versatile approach to antitumor therapy.[Mol Cancer Ther 2005;4(6):956–67]

IntroductionAntiangiogenic therapy designed to block neovasculariza-tion is an evolving anticancer strategy (1, 2). Antiangiogenictumor therapies have recently attracted intense interestbecause of their broad-spectrum of action, low toxicity, andabsence of drug resistance (1). Tumor angiogenesis isregulated by a balance of stimulators [vascular endothelialgrowth factor (VEGF), basic fibroblast growth factor(bFGF), etc.] and inhibitors of angiogenesis (angiostatin,endostatin, etc.). The efficacy of a humanized anti-VEGFantibody (bevacizumab, Avastin, recombinant humanizedmonoclonal antibody-VEGF; Genentech, South San Fran-cisco, CA) when used in combination with chemotherapyreported in a phase III trial in metastatic colon carcinomasuggests that antiangiogenic approaches may be used toaugment existing antitumor strategies (3–5).

Endostatin is a 20-kDa fragment of the a1 chain of typeXVIII collagen with potent antiangiogenic and antitumorproperties (6, 7). The antitumor activity of endostatin mayin part be due to inhibition of the proliferation andmigration of endothelial cells (8, 9). In addition, endostatinmay down-regulate VEGF expression in tumor cells (10).Systemic therapy with endostatin caused primary tumorregression in murine models without obvious toxicity (1, 2).Systemic endostatin also suppressed the growth of humanrenal cell cancer xenografts in a nude mouse model (11, 12).Antiangiogenic therapy with endostatin has been shown toblock tumor growth with no evidence for emergence ofresistance despite multiple cycles of therapy (2, 11). In somemurine models, repeated treatment with endostatin

Received 12/3/04; revised 2/25/05; accepted 4/13/05.

Grant support: Korea Research Foundation Grant in Korea, Departmentof Defense Breast Cancer Research Program grants BC044744 (DAMD17-05-1-0351) and NIH RO1 CA74273, University of Miami, SylvesterComprehensive Cancer Center, and Braman Family Breast Cancer Institute.

The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

Requests for reprints: Seung-Uon Shin, Department of Medicine,Hematology-Oncology, University of Miami School of Medicine andSylvester Comprehensive Cancer Center, 1475 Northwest 12th Avenue(D8-4), Miami, FL 33136. Phone: 305-243-3668; Fax: 305-243-4787.E-mail: [email protected]

Copyright C 2005 American Association for Cancer Research.

956

Mol Cancer Ther 2005;4(6). June 2005

on April 12, 2019. © 2005 American Association for Cancer Research. mct.aacrjournals.org Downloaded from on April 12, 2019. © 2005 American Association for Cancer Research. mct.aacrjournals.org Downloaded from on April 12, 2019. © 2005 American Association for Cancer Research. mct.aacrjournals.org Downloaded from

http://mct.aacrjournals.org/



resulted in permanent eradication of tumor (1, 2, 11, 13).These results show a potential advantage of antiangiogenictherapy because a major cause of treatment failure withcontemporary cytotoxic agents is the development ofacquired resistance (11).

In early human trials, endostatin administration at a doseof 240 mg/m2/d that were effective in tumor xenograftstudies did not produce significant changes in biologicalend points (14). ‘‘Modest’’ clinical benefit was observed in3 of 15 patients (14). In a phase I trial using dose levels upto 600 mg/m2/d, minimal antitumor activity was seen in 25patients despite circulating levels that had been effective inmouse models (15). Two of 25 patients (1 with sarcoma and1 with melanoma) showed antitumor activity. A thirdphase I trial of endostatin administered as a 1-hour i.v.infusion daily for a 28-day cycle at doses of 30 to 300 mg/m2

showed that endostatin was free of significant drug-relatedtoxicity, but no clinical responses were observed in all 21patients (16). These phase I clinical trials have proven thatendostatin is a very safe drug in a variety of doseschedules. However, these results did not show substantialantitumor activity. We believe that dosing and schedulesmay have been suboptimal due to the short half-life ofendostatin and/or that bulky disease in advanced andheavily pretreated patients may not be optimally respon-sive to recombinant human endostatin. Alternatively, thenumber of endostatin receptors may be diminished inestablished human tumors when compared with murinetumors, and/or endostatin-induced signaling may bedifferent in human tumors than in murine tumors resultingin different therapeutic outcomes.

The HER2 antigen encodes a receptor-like transmem-brane protein with a tyrosine kinase activity (17–19) and isexpressed at increased levels on several human cancers(30–40% of breast and ovarian cancers; ref. 20) and hasalready been successfully targeted in patients (Herceptin;humanized anti-HER2 IgG1; trastuzumab). In the presentstudy, to increase the efficacy of endostatin, we con-structed an anti-HER2 IgG3-CH3-endostatin fusion protein(aHER2-Endo) by joining murine endostatin to the 3V endof a humanized anti-HER2 IgG3. We show that, comparedwith normal endostatin, aHER2-Endo retains antiangio-genic activity, exhibits prolonged serum half-life andstability, selectively targets tumors bearing HER2, inhibitsblood vessel formation, and inhibits tumor growth moreeffectively in vivo than either endostatin or anti-HER2antibody alone or delivered in combination. Application ofaHER2-Endo to women with breast cancer will requirecareful evaluation of antibody fusion protein antigenicityand may benefit from use of a human endostatin fusiondomain.

Materials andMethodsCell Lines and AnimalsCT26, CT26-HER2 (21), human embryonic kidney 293,

and transfected Sp2/0 cells were cultured in Iscove’smodified Dulbecco’s medium (Cellgro, Mediatech, Inc.,

Herndon, VA) with 5% calf serum (Life Technologies-Invitrogen Corp., Carlsbad, CA). Female BALB/c mice(4–6 weeks) and severe combined immunodeficient (SCID)mice (4–6 weeks) were purchased from The JacksonLaboratory (Bar Harbor, ME). All animal experiments weredone in compliance with the NIH Guides for the Care andUse of Laboratory Animals and were approved by theUniversity of Miami Institutional Animal Care and UseCommittee.

Construction, Expression, and Characterization ofAHER2-Endo Fusion Protein

To obtain active endostatin, a mouse endostatin expres-sion vector (pFLAG-CMV-1-endostatin) was cotransfectedwith pcDNA3.1 (Clontech, Palo Alto, CA) into humanembryonic kidney 293 cells, and G418 (0.6 Ag/mL)–resistant cells were selected as described previously (22).Secreted endostatin was harvested from serum-free condi-tioned medium and purified in a heparin-Sepharose CL-6Bcolumn.

Experimental murine endostatin gene originated frompFLAG-CMV-1-endostatin by PCR using primers 5V-CCCCTCGCGATATCATACTCATCAGGACTTTCAGCCand 5V-CCCCGAATTCGTTAACCTTTGGAGAAAGAGGT-TCATGAAGC (22). PCR products were subcloned into p-GEM-T Easy Vector (Promega, Madison, WI). The subclonedendostatin gene was ligated in frame to the carboxyl end ofthe heavy chain constant domain (CH3) of human IgG3 inthe vector pAT135 as described previously (23) and theendostatin heavy chain constant region was then joined to ananti-HER2 variable region of a recombinant humanizedmonoclonal antibody 4D5-8 (HER2, trastuzumab; Genen-tech) in the expression vector (pSV2-his) containing HisDgene for eukaryotic selection (24, 25).

The aHER2-Endo fusion protein construct was stablytransfected into Sp2/0 myeloma cells stably expressing theanti-HER2 n light chain to assemble entire aHER2-Endofusion proteins as described previously (26). The aHER2-Endo fusion proteins were biosynthetically labeled with[35S]methionine (Amersham Biosciences, Piscataway, NJ)and analyzed by SDS-PAGE (26). The endostatin fusionprotein was purified from culture supernatants usingprotein A immobilized on Sepharose 4B fast flow (Sigma,St. Louis, MO).

To confirm that the endostatin moiety was present in theaHER2-Endo protein by Western blotting, rabbit anti-endostatin (BodyTech, Kangwon-Do, Korea) was used asthe primary antibody, and mouse anti-rabbit IgG conjugat-ed with horseradish peroxidase (Sigma) was used as thesecondary antibody. Goat anti-human IgG conjugated withhorseradish peroxidase was used to detect human anti-body.

ChorioallantoicMembraneAssayThe ability of aHER2-Endo to block VEGF/bFGF–

induced angiogenesis was tested by chorioallantoic mem-brane assay, which employed Leghorn chicken embryos(Charles River SPAFAS, Wilmington, MA) at 12 to 14 daysof embryonic development (12, 27). Vitrogen gel pellets(Collagen Biomaterials, Palo Alto, CA) were supplemented

Molecular Cancer Therapeutics 957


on April 12, 2019. © 2005 American Association for Cancer Research. mct.aacrjournals.org Downloaded from


with (a) vehicle (0.1% DMSO) in PBS alone (negativecontrol), (b) VEGF/bFGF (100 ng and 50 ng/pellet,respectively; positive control), or (c) VEGF/bFGF andeither aHER2-Endo, anti-HER2 IgG3, or endostatin atvarious concentrations (0.5, 1, 5, and 10 Ag/pellet) asdescribed (28). Polymerized mesh was placed onto theouter region of the chorioallantoic membrane of the embryoand incubated for 24 hours. To visualize vessels, FITC-dextran (400 AL; 100 Ag/mL; Sigma) was injected in thechick embryo bloodstream. Fluorescence intensity wasanalyzed with a computer-assisted image program (NIHImage 1.59). We calculated the two-tailed Student’s t testfor the overlapping concentrations tested for all proteins.

Pharmacokinetic and Biodistribution of AHER2-EndoAnti-HER2 IgG3 (100 Ag), aHER2-Endo (100 Ag), anti-

dansyl IgG3 (100 Ag), and endostatin (100 Ag) wereiodinated with 0.5 mCi 125I (Amersham Biosciences) bythe chloramine T method, which preferentially labelstyrosine groups (29). BALB/c mice (4–6 weeks) wereinjected s.c. with either 1 � 106 CT26-HER2 or CT26 cells orleft uninjected. Groups of three mice with either CT26-HER2 or CT26 tumors or no tumor were injected i.v. witheither 30 ACi [125I]aHER2-Endo, 32 ACi [125I]anti-HER2IgG3, 24 ACi [125I]endostatin, or 32 ACi [125I]anti-dansylIgG3. Blood samples were serially obtained at variousintervals ranging from 15 minutes to 96 hours from theretro-orbital plexus of mice injected with either aHER2-Endo, anti-HER2 IgG3, or anti-dansyl IgG3. Mice injectedwith endostatin alone were bled within 15 seconds to 60minutes after the i.v. injection.

To monitor stability of the endostatin fusion protein, theiodinated fusion proteins in serum were resolved throughSDS-PAGE and compared by molecular size with theiodinated antibody controls. The trichloroacetic acid–precipitable radioactivity in each blood sample wasmeasured in a gamma counter. The pharmacokineticvariables were calculated by fitting plasma radioactivitydata to a biexponential model with derivative free nonlinearregression analysis (PARBMDP, Biomedical Computer PSeries Program developed at University of California at LosAngeles Health Sciences Computing Facilities). The datawere weighed using weight = 1 / (concentration)2, whereconcentration was expressed as either counts per minute(cpm) per microliter (AL) or percentage of injected dose(%ID) per milliliter. The pharmacokinetic variables, such asplasma clearance, initial plasma volume, systemic volumeof distribution, steady-state area under the plasma concen-tration curve (AUC0-1), and mean residence time, werecalculated from the slopes and intercept of the biexponen-tial equation as described previously (30).

Following the pharmacokinetic experiments, mice wereex-sanguinated by perfusion with 20 mL PBS for measure-ments of the tissue distribution of 125I-labeled antibody-endostatin fusion protein. The heart, lung, liver, spleen,kidney, muscle, and tumor were removed, weighed, andgamma counted and the %ID/g of tissue was calculated.Specific tumor targeting is expressed as the radiolocaliza-tion index (the %ID/g in tumor divided by the %ID/g in

blood). To determine the distribution and localization ofthe 125I-labeled proteins in mice simultaneously implantedwith CT26 and CT26-HER2 tumors on opposite flanks,groups of three mice were injected i.v. with either 5 ACi[125I]aHER2-Endo fusion protein or 5 ACi [125I]anti-HER2IgG3. The animals were sacrificed at different times (6, 24,and 96 hours) after injection of labeled protein, and organs(e.g., lung, liver, kidney, spleen, muscle, CT26 tumor, CT26-HER2 tumor, blood, and urine) were isolated afterperfusion of the mouse with PBS, weighed, and countedin a gamma scintillation counter. The %ID/g for each organwas determined as above.

In vivo Antitumor EffectsMurine colon adenocarcinoma CT26 cells were trans-

duced with the gene for HER2 antigen as describedpreviously (21). The CT26-HER2 cells that were used inthese studies proliferate at the same rate in vitro as parentalCT26 cells (data not shown). The in vivo antitumor efficacyof aHER2-Endo was examined using the CT26 or CT26-HER2 cell lines implanted in the syngeneic BALB/c mice.To determine targeting and efficacy of aHER2-Endo,BALB/c mice (8 per group, 4–6 weeks) were injected s.c.with 1 � 106 CT26-HER2 cells in the right flank and/orcontrol CT26 cells in the left flank. On day 7, mice (8 pergroup) were injected i.v. with the aHER2-Endo fusionproteins (42 Ag/injection, 2 � 10�10 mol, equimolar to 8 Agendostatin), anti-HER2 IgG3 alone (34 Ag/injection, 2 �10�10 mol), endostatin alone (8 Ag/injection, 4 � 10�10 mol),or the combination of anti-HER2 IgG3 (34 Ag) and endo-statin (8 Ag) or PBS as a control. All mice received seveninjections delivered at 2-day intervals. Tumor size wasmeasured with calipers and growth rates were recordedand calculated using the following equation: tumor volume(mm3) = 4 / 3 � 3.14 � {(long axis + short axis) / 4}3.

The human breast cancer SK-BR-3 xenograft model inSCID mice was also used to evaluate antitumor activity ofaHER2-Endo fusion protein. SK-BR-3 cells (1 � 106 permouse) were implanted s.c. in the flank of SCID mice. Onday 15, mice (8 per group) were injected i.v. with theaHER2-Endo fusion proteins (42 Ag), anti-HER2 IgG3(34 Ag), the combination of anti-HER2 IgG3 (34 Ag) andendostatin (8 Ag), or endostatin (8 Ag). This treatment wasrepeated every other day for 11 doses. Tumor growth wasanalyzed as described above.

Immunohistochemistry and Image Analysis of BloodVessel Formation

Mice were killed at the end of the experiments and frozentumors were stored at �80jC until further use. Forconventional immunohistochemistry, 5 Am tissue sectionswere cut using a cryostat (Shandon, Pittsburgh, PA) andplaced on positively charged slides (Fisher Scientific,Pittsburgh, PA; ref. 31). To analyze the microvesselformation in tumors, sections were stained with a ratanti-mouse platelet-endothelial cell adhesion molecule-1(CD31) monoclonal antibody (PharMingen, San Diego, CA)and subsequently with the avidin-biotin complex method(Vector Laboratories, Burlingame, CA). HER2 expressionon tumors has been examined by staining tumor sections

Antibody-Endostatin Fusion Protein958




with anti-HER2 IgG3. All sections were counterstainedwith hematoxylin (Sigma). Positively stained vascularendothelial cells (brown) were visualized and imagedusing a digital camera attached to a Zeiss microscope (CarlZeiss, Thornwood, NY).

For confocal microscopic analysis, 30 Am cryosectionswere cut and stained with a rat anti-mouse CD31monoclonal antibody (31–33). Blood vessels were visual-ized with anti-rat IgG-Alexa 594 (Molecular Probes,Eugene, OR). Fluorescent blood vessels were then viewedvia LSM5 confocal microscope (Carl Zeiss), and 14 to 21digital images were obtained per section. These digitalimages have been composed as one image per each sectionto measure blood vessel density, and blood vessel area(pixel2) was then computed from the composite images andaveraged to measure blood vessel density per tumor.Images were analyzed using NIH ImageJ version 1.31software by color to form a binary image of the tumorblood vessels.

Statistical AnalysisAntiangiogenic activity, pharmacokinetics, biodistribu-

tion, and tumor growth are presented as the mean F SE.Statistical analysis was done using ANOVA and Student’st test. Differences were considered statistically significantat P < 0.05.

ResultsProduction and Antiangiogenic Activity of AHER2-

EndoThe aHER2-Endo fusion protein construct was stably

transfected into Sp2/0 myeloma cells and the secreted[35S]methionine-labeled protein has a molecular weight off220 kDa under nonreducing conditions (data notshown), the size expected for a complete antibody(170 kDa) with two molecules of endostatin (25 kDaeach) attached (Fig. 1A). Following reduction, heavy andlight chains of the expected molecular weight wereobserved (data not shown). We confirmed the presenceof endostatin in the fusion protein by Western blottingwith a rabbit anti-endostatin antiserum and goat anti-

human IgG (Fig. 1B). aHER2-Endo was identified at themolecular weight of 220 kDa by both anti-human IgGand anti-endostatin antibody. Following reduction, theheavy chain band from aHER2-Endo migrated at theexpected size of 85 kDa (data not shown). Integrity and

Figure 1. Characterization of aHER2-Endo. A, schematic diagram ofaHER2 IgG3 and aHER2-Endo. B, Western blot analysis of aHER2-Endo.The purified aHER2-Endo, anti-dansyl IgG3, and endostatin were resolvedunder nonreducing conditions and transferred onto a nylon membrane. Todetect the endostatin moiety, rabbit anti-endostatin was used as a primaryantibody, and mouse anti-rabbit IgG conjugated with horseradish perox-idase was used as a secondary antibody. Goat anti-human IgG conjugatedwith horseradish peroxidase was used to detect human antibody. C,inhibition of the angiogenic response induced by VEGF/bFGF. PurifiedaHER2-Endo preparation 1 (o) and preparation 2 (.) were added to analiquot of Vitrogen supplemented with a combination of VEGF and bFGF,and the mixture was placed on a nylon mesh. Impregnated mesh were thenplaced on the chick embryo and incubated as described in Materials andMethods. New vessel growth was visualized with FITC-dextran andmeasured by fluorescent intensity. Anti-HER2 IgG3 (E) and endostatin (4)are included for comparison. Positive control group (n) contains VEGF/bFGF alone, and negative control group (5) contains only vehicle. Points,mean (n = 5); bars, SE.





purity of aHER2-Endo protein purified using protein Aaffinity columns was tested under nonreducing condi-tions by SDS-PAGE (data not shown). aHER2-Endo wasidentified as a single band at the molecular weight of 220kDa (data not shown).

The ability of endostatin to block VEGF/bFGF–inducedangiogenesis in vitro was tested using the chorioallantoicmembrane assay. Pellets containing Vitrogen and VEGF/bFGF (100 ng and 50 ng/pellet, respectively) and eitheranti-HER2 IgG3 (0.5, 1, 5, or 10 Ag/pellet: 2.95, 5.9, 29.5, or59 pmol/pellet), aHER2-Endo (0.5, 1, 5, or 10 Ag/pellet:2.25, 4.5, 22.5, or 45 pmol/pellet), or endostatin (0.5, 1, 5, or10 Ag/pellet: 20, 40, 200, or 400 pmol/pellet) weremeasured for invasion by newly formed capillaries(Fig. 1C). Two independent preparations of anti-HER2antibody-endostatin fusion protein were able to suppressthe angiogenic response mediated by VEGF/bFGF in adose-dependent manner with a specific activity similar tothat seen with endostatin (P = 0.566 and 0.516; Fig. 1C) andshowed significantly better antiangiogenic activity thananti-HER2 IgG3 (P = 0.034 and 0.050; Fig. 1C). Therefore,genetically engineered aHER2-Endo is able to inhibit theangiogenic response mediated by VEGF/bFGF, similar tonative endostatin.

Serum Clearance and Stability of AHER2-EndoTo characterize the pharmacokinetics of aHER2-Endo,

mice with or without implanted tumors (CT26 or CT26-HER2) were injected i.v. with [125I]aHER2-Endo, anti-HER2IgG3, endostatin, or a control anti-dansyl IgG3 andclearance of injected radiolabeled proteins was measured.Representative results from mice implanted with CT26-HER2 tumors are shown graphically in Fig. 2, whereasnormal mice were used as a control and the pharmacoki-netic data for all mice are summarized in Table 1.[125I]endostatin was rapidly removed from the plasmacompartment in mice with or without tumors (T1/2

2

elimination, 0.6–3.8 hours), whereas the clearance rate of[125I]aHER2-Endo (T1/2

2, 40.2–44.0 hours) was similar orslightly increased to that of [125I]anti-HER2 IgG3 (T1/2

2,39.9–63.0 hours) and anti-dansyl IgG3 (T1/2

2, 43.7–46.5hours; Fig. 2A; Table 1). Therefore, endostatin fused withantibody has a serum half-life of at least 10-fold greaterthan endostatin alone.

Although the AUC of aHER2-Endo (218% ID h/mL) wasreduced f6-fold compared with anti-HER2 IgG3 (1,243%ID h/mL) in mice bearing CT26-HER2 tumors (Table 1),AUC of aHER2-Endo was increased by a factor of 56compared with endostatin (3.9% ID h/mL) as a conse-quence of both a longer half-life of elimination (70-foldincrease: 44.0 versus 0.63 hours) and an increased meanresidence time (56-fold increase: 46.7 versus 0.83 hours).Endostatin was very rapidly removed from serum within30 minutes, principally by glomerular filtration and renalclearance, but aHER2-Endo showed much slower clearancefrom serum than endostatin, albeit slightly faster thanclearance that seen for anti-HER2 IgG3 and anti-dansylIgG3. The difference in AUC between aHER2-Endo andantibodies (anti-HER2 IgG3 and anti-dansyl IgG3) might be

due to the endostatin domain in fusion proteins, whichmay cause preferential deposition in tissues expressinghigh levels of endostatin-binding proteins, such as the av

and a5 integrins, and/or proteoglycans containing heparansulfate (34, 35).

To analyze the serum stability of 125I-labeled anti-HER2IgG3, aHER2-Endo, endostatin, and anti-dansyl IgG3,

Figure 2. Serum clearance and stability in mice bearing CT26-HER2tumors. Serum clearance (A) and serum trichloroacetic acid (TCA )precipitability (B) of 125I-labeled aHER2-Endo (.), anti-dansyl IgG3 (5),anti-HER2 IgG3 (n), and endostatin (o) were measured. Measurements ofanti-HER2 IgG3 and aHER2-Endo were made up to 96 h after i.v. injectionand those of endostatin up to 60 min. Points, mean (n = 3 BALB/c mice);bars, SE. Statistical analysis of serum stability at 5 min to 1 h was doneusing Student’s t test (paired, two-tailed distribution). Serum samples of125I-labeled proteins were analyzed by SDS-PAGE (C). Each iodinatedinitial protein was used as an initial control for its own serum samples. 125I-labeled anti-HER2 IgG3 was also used as a control.





plasma samples were trichloroacetic acid precipitated andcounted (Fig. 2B). Ninety-six hours following injection,f90% of the anti-HER2 IgG3 and aHER2-Endo in serumremained intact. For endostatin, f90% were intact 2minutes after injection and only 55% remained intact at60 minutes. aHER2-Endo cleared much more slowly withkinetics resembling anti-HER2 IgG3 and anti-dansyl IgG3.Analysis of serum samples by SDS-PAGE confirmed thataHER2-Endo in circulation remained intact (Fig. 2C).

Thus, the endostatin moiety of aHER2-Endo is signifi-cantly more stable in the circulation than endostatin (P =0.0014, comparing 5-minute time point with 1-hour timepoint).

Biodistribution and Biolocalization of AHER2-EndoTo measure localization and biodistribution, we ana-

lyzed the tumor/blood ratio of injected radiolabeledprotein over time. Ninety-six hours after an i.v. injectioninto mice bearing CT26-HER2 tumors, anti-HER2 IgG3

Table 1. Pharmacokinetic variables for 125I-labeled proteins in mice with or without tumors

Mice Variable* Anti-dansyl IgG3 Anti-HER2 IgG3 aHER2-Endo Endostatinc

No tumorb T1/21 (h): distribution 0.70 F 0.44 0.48 F 0.06 0.028 F 0.004

T1/22 (h): elimination 39.9 F 18.3 40.2 F 2.9 3.75 F 2.18

AUC0-48 (%ID h/mL) 658 F 113 389 F 87 3.3 F 0.4AUC0-1 (%ID h/mL) 1,474 F 593 667 F 135 18.2 F 8.5Mean residence time (h) 54.8 F 24.2 56.5 F 3.8 5.25 F 2.75

CT26 T1/21 (h): distribution 2.62 F 1.07 1.61 F 0.21 10.33 F 8.47 0.028 F 0.004

T1/22 (h): elimination 43.7 F 2.2 62.2 F 3.8 43.3 F 22.7 1.10 F 0.10

AUC0-96 (%ID h/mL) 847 F 165 1,157 F 60 358 F 76 2.8 F 0.2AUC0-1 (%ID h/mL) 1,063 F 247 1,717 F 88 470 F 109 5.4 F 0.6Mean residence time (h) 56.0 F 6.2 85.3 F 4.62 66.0 F 16.2 1.44 F 0.14

CT26-HER2/neu T1/21 (h): distribution 15.73 F 8.52 4.93 F 1.17 3.37 F 0.95 0.014 F 0.007

T1/22 (h): elimination 46.5 F 22.2 63.0 F 6.7 44.0 F 4.0 0.63 F 0.29

AUC0-96 (%ID h/mL) 1,083 F 123 887 F 24 185 F 17 3.0 F 0.7AUC0-1 (%ID h/mL) 1,453 F 242 1,243 F 50 218 F 20 3.9 F 0.3Mean residence time (h) 67.7 F 12.0 76.3 F 10.1 46.7 F 10.6 0.83 F 0.35

*For the pharmacokinetic variables, the superscript 1 represents the distribution phase and the superscript 2 represents the elimination phase. AUC0-96 andAUC0-1 are the first 96 hours and steady-state AUC, respectively. To calculate pharmacokinetic variables, the plasma radioactivity results were fit to abiexponential model (endostatin, anti-HER2 IgG3, and aHER2-Endo) with a derivative-free nonlinear regression analysis. Data are mean F SE (n = 3 BALB/cmice).cMeasurements of endostatin were made 1 hour after i.v. injection in the mice.bMeasurements of 125I-labeled proteins in mice without tumors were made 48 hours after i.v. injection in the mice.

Figure 3. Targeting of anti-HER2IgG3 and aHER2-Endo to CT26-HER2tumors, CT26 tumors, or other organsin BALB/c mice. A, two groups ofBALB/c mice (12 per group) wereinjected s.c. with 106 single-cell sus-pensions of either CT26-HER2 (blackcolumns ) or CT26 (white columns).When the tumors were f5 mm indiameter, 125I-labeled target proteinswere injected into four groups of mice(n = 3 per protein) through the tailvein. Specific tumor targeting isexpressed as the radiolocalization in-dex (the%ID/g in tumor divided by the%ID/g in blood). B–D, CT26 (C) andCT26-HER2 (D) were contralaterallyimplanted within the same BALB/cmice (B; n = 3 per group) andindicated iodinated proteins wereinjected. Following injection, the indi-cated tissues were harvested at thetimes indicated after injection (6, 24,and 96h) and%ID/gwasmeasured asoutlined in Materials and Methods.Specific tumor targeting is expressedas %ID/g of tissue. Columns, mean;bars, SE.





was found mainly in the tumor and blood (5.67% and2.10% ID/g, respectively; data not shown). The radio-localization indices at 96 hours after injection (the %ID/gin tumor divided by the %ID/g in blood) of aHER2-Endo and anti-HER2 IgG3 were similar (Fig. 3A).aHER2-Endo showed a tumor/blood ratio of 3.76 forCT26-HER2 and 0.50 for CT26, whereas anti-HER2 IgG3showed tumor/blood ratios of 2.83 and 0.47 for CT26-HER2 and CT26, respectively (Fig. 3A). Therefore, bothanti-HER2 antibody and anti-HER2 antibody-endostatinfusion protein preferentially localized to HER2-expressingtumors.

To directly measure localization of antibody-endostatinfusion proteins to the antigenic target, mice simulta-neously implanted with CT26 and CT26-HER2 tumors onopposite flanks were injected i.v. with either 125I-labeledaHER2-Endo or 125I-labeled anti-HER2 IgG3 (Fig. 3). Thebiodistribution and biolocalization of the labeled proteinswas examined at different times (6, 24, and 96 hours)after injection of labeled proteins (Fig. 3B). aHER2-Endoand anti-HER2 IgG3 preferentially localized to CT26-HER2 tumors (Fig. 3C and D). Specific tumor radio-localization indices (the %ID/g in CT26-HER2 tumordivided by the %ID/g in CT26) of aHER2-Endo (5.34,7.42, and 3.55 at 6, 24, and 96 hours, respectively) wereconsistently equal to or greater than those of anti-HER2IgG3 (1.12, 2.12, and 2.60, respectively; Fig. 3C and D).This indicates that the relative localization of targetedantibody-endostatin fusions to tumor is largely due tobinding to the HER2 target antigen, and targeting

functions of the antibody are well preserved. Whetherthe endostatin domain also contributes to the localizationinto tumor we observed is not known, because endostatindid not preferentially localize to HER2-expressing tumorin our hands (Fig. 3A).

AntitumorActivity of AHER2-Endo In vivoPreliminary experiments revealed that the CT26-HER2

tumors implanted in BALB/c mice grew at the same rateas the parental CT26 tumors. HER2 expression wasmaintained in CT26-HER2 implanted into BALB/c micefor up to 1 month following implantation as determinedby immunohistochemistry (21). In initial experimentsusing the CT26-HER2 model, aHER2-Endo administrationinhibited tumor growth more efficiently than either anti-HER2 IgG3 or endostatin alone (data not shown). Todetermine whether targeting to HER2 antigen wasimportant, mice were simultaneously implanted withCT26 and CT26-HER2 on opposite flanks. Administrationof aHER2-Endo showed preferential inhibition of CT26-HER2 compared with contralaterally implanted CT26parental tumor (Fig. 4). aHER2-Endo inhibited moreeffectively than anti-HER2 IgG3 antibody, endostatin, orthe combination of antibody and endostatin (P = 0.041,0.041, and 0.022, respectively). These results suggest thattargeting to HER2 antigen may have increased efficacy ofthe fusion protein.

We next determined whether aHER2-Endo, endostatin,anti-HER2 IgG3 antibody, or both antibody and endostatinin combination would inhibit the growth of SK-BR-3human breast cancer xenografts in SCID mice. SK-BR-3

Figure 4. Antitumor activity ofaHER2-Endo fusion protein. A–C,syngeneic mouse model. BALB/cmice (C; n = 8 per group) wereimplanted s.c. contralaterally withCT26 (A) and CT26-HER2 (B; 1 �106 cells per mouse) followed on day7 by equimolar injections every otherday (arrows , seven times) of PBS (o),aHER2-Endo (.), anti-HER2 IgG3 (5),endostatin (n), or the combination ofanti-HER2 IgG3 and endostatin (E) asindicated in Materials and Methods.D, SCID mice model bearing humanbreast cancer SK-BR-3 xenografts.SCID mice (n = 8 per group) wereimplanted s.c. with SK-BR-3 (1 � 106

cells per mouse). On day 15, equimo-lar amounts of aHER2-Endo (.), anti-HER2 IgG3 (5), endostatin (n), andcombination of anti-HER2 IgG3 andendostatin (E) or PBS (o) wereinjected every other day (arrows , 10times). Points, mean tumor measure-ments; bars, SE. Statistical analysiswas done using ANOVA and Stu-dent’s t test (paired, two-tailed distri-bution).





was implanted on the flank of SCID mice. Repeatedadministration of aHER2-Endo resulted in a significantlygreater reduction of tumor volume compared with eitheranti-HER2 antibody alone, endostatin alone, or antibodyand endostatin given in combination (P = 0.010, 0.046, and0.038, respectively; Fig. 4D).

Blood Vessel Formation in CT26-HER2 Tumors Trea-tedwith the AHER2-Endo Fusion Protein

Mice were simultaneously implanted with CT26 andCT26-HER2 tumors on opposite flanks and tumors allowedto grow to a diameter of 4 to 6 mm at which time the micewere i.v. treated with aHER2-Endo fusion proteins. CT26-HER2 tumors grew more slowly in mice treated withaHER2-Endo compared with control mice, and kineticswere similar to those shown in Fig. 4 (data not shown).After five treatments, tumors were removed and cryosec-tions were immunohistochemically stained for endothelialcells with anti–platelet-endothelial cell adhesion molecule-

1 (CD31) antibody or for HER2 expression with anti-HER2antibody (Fig. 5A–G). CT26-HER2 tumors continued toexpress HER2 on their surface (Fig. 5F and G). CT26 tumorsand CT26-HER2 tumors treated with PBS or anti-HER2IgG3 showed a larger number of vessels by immunohisto-chemistry than CT26-HER2 treated with endostatin fusionproteins (Fig. 5A–E).

To characterize and quantitate blood vessel numberand density, tumor sections were stained with rat anti-mouse anti–platelet-endothelial cell adhesion moleculeantibody and staining detected with anti-rat IgG-Alexa 594(Fig. 5H–J). Confocal microscopic analysis revealedstriking differences in the extent and organization ofthe vasculature in treated tumors. Composites of 14 to 21digital microscopic images showed that blood vessels inCT26-HER2 tumors treated with aHER2-Endo were lessorganized and contained fewer branching capillariescompared with control tumors (Fig. 5H and I). The

Figure 5. Analysis of tumor vascularity. A–G, immunohistochemical staining of blood vessels in treated or untreated CT26 and CT26-HER2 tumors.Cryosections of CT26 and CT26-HER2 tumors with or without treatment were stained with anti-CD31 antibody for endothelial cells (A–E) or anti-HER2antigen (F and G). CT26 tumor (A, B, and F), CT26-HER2 tumor (C–E and G), no treatment (PBS; A and C), treatment with aHER2-Endo (B and D),treatment with anti-HER2 IgG3 (E). Magnification, �400. H and I, visualization of blood vessel formation in representative CT26-HER2 tumor sections.Tumor sections were prepared from CT26-HER2 tumors without treatment (PBS; H) or CT26-HER2 tumors treated with aHER2-Endo (I). Each cryosectionwas stained with rat anti-mouse CD31 and anti-rat IgG-Alexa 594 (red fluorescence). Fourteen to 21 digital images (magnified �400) were obtained persection, and the above images are composite figures. J, quantification of blood vessel area in CT26 and CT26-HER2 tumors. The composite images wereanalyzed using NIH ImageJ version 1.31 by color image to form a binary image to allow measurement of blood vessel density. Blood vessel area (pixel2)was then computed. Columns, mean; bars, SE. Statistical comparison was done using Student’s t test (paired, two-tailed distribution).





blood vessel density was measured by determining thearea that was occupied by vessels as described inMaterials and Methods. CT26-HER2 tumors treated withendostatin fusion had significantly less vascular area(16% of untreated CT26-HER2; P = 0.028) than did theuntreated CT26-HER2 tumors (Fig. 5J). Vessel densitywas also markedly reduced in the CT26-HER2 tumorstreated with aHER2-Endo relative to parental CT26 oruntreated CT26-HER2 tumors (Fig. 5J).

DiscussionDespite promising results in murine studies involvingangiogenic inhibition by endostatin, results in preliminaryclinical trials in man have been disappointing. The reasonfor the lack of efficacy are not known but may in partrelate to the short endostatin half-life and poor stabilityof endostatin. The short in vivo half-life and the seruminstability of endostatin necessitate frequent administra-tion and may decrease clinical efficacy. It is possible thatsustained levels of endostatin will be required toeffectively inhibit tumor growth. Indeed, patients withDown syndrome show a remarkably low incidence ofsolid tumors. This has been attributed by some inves-tigators to circulating endostatin levels approximatelytwice those seen in normals (36). This suggests thatmodest but sustained increases in endostatin may becapable of significantly inhibiting tumor growth. Al-though it may be possible to improve therapy by usingcontinuous or higher dosage, this increases the risk ofsystemic side effects on normal physiologic processesinvolving vascular repair or vessel outgrowth, such asthe angiogenic response to ischemia and wound healing(37, 38).

To produce a more effective form of trastuzumab andimprove the efficacy of endostatin, we constructed anaHER2-Endo fusion protein by joining murine endostatinto the CH3 terminus of the antitumor antibody, anti-HER2 IgG3. We hypothesized that several of thelogistical disadvantages of the long-term treatment withhigh dosages of endostatin could be overcome if thehalf-life and stability of endostatin could be increasedand if endostatin could be specifically targeted to thetumor to achieve higher local concentrations and greaterspecificity. In this report, we show that the serum half-life and stability of endostatin were dramatically in-creased by fusion to an anti-HER2 antibody. Both theantiangiogenic properties of endostatin and the tumortargeting function of the antibody were fully maintainedin the fusion protein. The fusion protein selectivelytargets tumors that express the target antigen HER2,increasing the local delivery of endostatin and decreasingtumor growth. In addition, reduction of tumor growthby treatment with the endostatin fusion protein corre-lates with a reduction in tumor blood vessel density andcomplexity (Fig. 5).

Endostatin fused with antibody was cleared from theperipheral compartment much more slowly than free

endostatin (Table 1). The antibody moiety of the endo-statin fusion protein prolonged the serum half-life of thefused endostatin. A similar effect has been observed withother antibody-fusion proteins, including antibody-avidin(23, 39). The human prolactin antagonist G129R-endo-statin fusion protein also had a prolonged serum half-lifecompared with that of G129R or endostatin and showedaugmented antitumor effects on a murine mammaryadenocarcinoma 4T1 in a nude mouse tumor model (40).The longer half-life and ability to target tumors ofaHER2-Endo might render it effective at lower and lessfrequent doses.

To achieve high concentration of endostatin in circula-tion, transfer of a retroviral vector encoding a secretableendostatin into hematopoietic stem cells resulted high-leveland long-term secretion of endostatin in hematopoieticstem cell–transplanted mice. However, no inhibition ofneoangiogenesis or the growth of primary or metastaticT241 fibrosarcoma was observed (41). In addition, elevatedconcentrations of endostatin also did not inhibit the growthof human B-lineage acute lymphocytic leukemia in severalanimal models (42). The hematopoietic stem cell trans-plants might not work in mice because of instability ofendostatin in circulation. Circulating human endostatin hasbeen isolated from patient blood with chronic renalinsufficiency, but this endostatin showed no antiprolifer-ative effect on bovine microcapillary endothelial cells orhuman umbilical vascular endothelial cell following bFGFstimulation (43). Circulating endostatins may lose anti-angiogenic activity over time (43). In the present study,endostatin has also been shown to be unstable in serumand was rapidly eliminated (Fig. 2B and C). In contrast,local expression of endostatin using viral vectors in thetumor vicinity did inhibit the growth of tumor in severalmouse models (38, 44–47). However, it is impractical tolocally inject metastases especially in man. Viral vectorsmay cause inflammation and elicit both an innate and anadaptive immunologic response on repeated injection. It isalso difficult to titrate dose reproducibly. Thus, geneticdelivery of endostatin is likely to be difficult to regulate invivo and is unlikely to result in reliable and reproduciblelevels in a manner suitable for clinical application (48–50).Local delivery using an antibody fusion protein as atargeting vehicle may allow increased local concentrationsand more reliably target microscopic tumor deposits.Because f90% of aHER2-Endo in serum remained intact96 hours following injection (Fig. 2B and C), the increasedstability and/or local delivery of aHER2-Endo may renderit a more effective antiangiogenic agent compared withendostatin.

Both anti-HER2 antibody and aHER2-Endo preferentiallylocalized to HER2-bearing tumors, whereas endostatin andanti-dansyl antibody failed to localize to tumor in ourstudies (Fig. 3A). In vivo localization of anti-HER2 antibodyand aHER2-Endo to tumor is due to binding to HER2target antigen. Consequently, treatment with aHER2-Endopreferentially inhibited growth of CT26-HER2 comparedwith parental CT26 tumors (Fig. 4). aHER2-Endo inhibited





more effectively than the equimolar combination ofantibody and endostatin. These data suggested thatcombining the targeting capability of anti-HER2 antibodywith the antiangiogenic activity of endostatin improved theantitumor activity of both agents more than that whichcould be achieved by simultaneous administration of bothagents.

The targeting of antiangiogenic proteins using antibodyfusion proteins may be highly selective for tumors andrelatively nontoxic compared with other strategies, suchas the attachment of radioligands (e.g., Zevalin or Bexxar)or chemotherapeutic agents (e.g., gemtuzumab ozogami-cin; Mylotarg; anti-CD33 antibody conjugated withcalicheamicin; refs. 51–53). Because the overall responserates of HER2+ breast cancers to trastuzumab remainrelatively low (15–34%; refs. 54–56), this approach holdspromise for increasing both response rate and durabilityrelative to trastuzumab.

Several different mechanisms may account for theantitumor activity of trastuzumab. These include thedown-regulation of HER2 in some tumors, resulting inthe subsequent inhibition of its downstream phosphatidy-linositol 3-kinase-Akt signaling pathway (57, 58) and theinduction of G1 arrest and the cyclin-dependent kinaseinhibitor p27 (59). Trastuzumab may also indirectly affectangiogenesis. VEGF expression in some primary HER2+

tumors may be decreased by trastuzumab (60–63). Morerecently, investigators have shown 4- to 5-fold induction ofthe antiangiogenic factor thrombospondin-1 in trastuzu-mab-treated patients (63). This suggests that furtheraugmentation of antiangiogenic properties might furtherenhance trastuzumab efficacy.

Trastuzumab may also inhibit growth by inducingantibody-dependent cell-mediated cytotoxicity and/orcomplement-dependent cytotoxicity (64, 65). In our hands,aHER2-Endo and anti-HER2 IgG3 showed similar abilityto mediate antibody-dependent cell-mediated cytotoxicity(data not shown).aHER2-Endo also inhibited the growth of the human

breast cancer SK-BR-3 xenografts in SCID mice andreduced tumor growth more than either anti-HER2antibody alone, endostatin alone, or antibody and endo-statin in combination (Fig. 4D). This inhibition seems toinvolve both anti-HER2 antibody and endostatin activities,because anti-HER2 IgG3 alone also had significantantitumor activity in this model. Antitumor activity oftrastuzumab is largely restricted to tumors with highlevels of HER2 overexpression and/or HER2 geneamplification, and the human breast tumor SK-BR-3 cellline expresses high levels of HER2 (f43 HER2 genecopies per cell; ref. 66). Unlike SK-BR-3, CT26-HER2tumors do not require HER2 signaling for proliferation(58–61) and anti-HER2 IgG3 showed minimal activity.Therefore, the enhanced antitumor activity of aHER2-Endo in CT26-HER2 tumors is likely due to theantiangiogenic activity of the endostatin domain. Inin vivo targeting experiments (Fig. 3A), aHER2-Endolocalized slightly better to the CT26-HER2 than anti-

HER2 IgG3. Although the difference in targeting ofaHER2-Endo and anti-HER2 IgG3 on CT26-HER2 tumorswas not statistically significant (P = 0.406) and the ratio oftumor/blood may not be constant over time, the possi-bility that the slightly better targeting of aHER2-Endomight enhance the antitumor activity of the fusion proteincannot be excluded. Although selective targeting ofendostatin to tumor has been reported (67), that did notseem to be the case for CT26.

Our studies suggest that specific targeting of theantiangiogenic agent and the prolongation of endostatinhalf-life and stability result in targeted suppression ofangiogenesis within the tumor bed. Linking endostatin toan antibody may therefore provide a unique opportunity toenhance the spectrum of trastuzumab activity and a moreeffective means of delivering endostatin. Application of thestrategy to women with breast cancer will require carefulevaluation of antibody fusion protein antigenicity and maybenefit from use of a human endostatin fusion domain.Thus, aHER2-Endo should be humanized to reduce thepossible antigenicity of the murine endostatin domain.Targeting antiangiogenic proteins using antibody is apotentially versatile approach that could be applied toother tumor and antibody targets (such as epidermalgrowth factor receptor or prostate-specific membraneantigen) as well. Other antiangiogenic domains could alsopotentially be used (angiostatin, tumstatin, etc.). Thepotential use of aHER2-Endo in combination with othertherapies, such as chemotherapy and/or other antiangio-genic approaches, in the setting of both minimal residualand/or metastatic disease may also further improve resultsand should be further investigated.

Acknowledgments

We thank Hae-Jung Kim for technical support and Dr. Lossos for statisticalanalysis.

References

1. Boehm T, Folkman J, Browder T, O’Reilly MS. Anti-angiogenic therapyof experimental cancer does not induce acquired drug resistance. Nature1997;390:404–7.

2. O’Reilly MS, Boehm T, Shing Y, et al. Endostatin: an endogenousinhibitor of angiogenesis and tumor growth. Cell 1997;88:277–85.

3. Willett CG, Boucher Y, di Tomaso E, et al. Direct evidence that theVEGF-specific antibody bevacizumab has antivascular effects in humanrectal cancer. Nat Med 2004;10:145–7.

4. Hurwitz H, Fehrenbacher L, Novotny W, et al. Bevacizumab plusirinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer.N Engl J Med 2004;350:2335–42.

5. Kabbinavar F, Hurwitz HI, Fehrenbacher L, et al. Phase II, randomizedtrial comparing bevacizumab plus fluorouracil (FU)/leucovorin (LV) withFU/LV alone in patients with metastatic colorectal cancer. J Clin Oncol2003;21:60–5.

6. Oh SP, Warman ML, Seldin MF, et al. Cloning of cDNA and genomicDNA encoding human type XVIII collagen and localization of the a1(XVIII)collagen gene to mouse chromosome 10 and human chromosome 21.Genomics 1994;19:494–9.

7. Rehn M, Pihlajaniemi T. a1(XVIII), a collagen chain with frequentinterruptions in the collagenous sequence, a distinct tissue distribution,and homology with type XV collagen. Proc Natl Acad Sci U S A 1994;91:4234–8.





8. O’Reilly MS, Holmgren L, Chen C, Folkman J. Angiostatin induces andsustains dormancy of human primary tumors in mice. Nat Med 1996;2:689–92.

9. Eriksson K, Magnusson P, Dixelius J, Claesson-Welsh L, Cross MJ.Angiostatin and endostatin inhibit endothelial cell migration in response toFGF and VEGF without interfering with specific intracellular signaltransduction pathways. FEBS Lett 2003;536:19–24.

10. Hajitou A, Grignet-Debrus C, Devy L, et al. The anti-tumoral effect ofendostatin and angiostatin is associated with a down-regulation ofvascular endothelial growth factor expression in tumor cells. FASEB J2002;16:1802–4.

11. Kerbel RS. A cancer therapy resistant to resistance. Nature 1997;390:335–6.

12. Dhanabal M, Ramchandran R, Volk R, et al. Endostatin: yeastproduction, mutants, and anti-tumor effect in renal cell carcinoma. CancerRes 1999;59:189–97.

13. Kisker O, Becker CM, Prox D, et al. Continuous administration ofendostatin by intraperitoneally implanted osmotic pump improves theefficacy and potency of therapy in a mouse xenograft tumor model.Cancer Res 2001;61:7669–74.

14. Eder JP Jr, Supko JG, Clark JW, et al. Phase I clinical trial ofrecombinant human endostatin administered as a short intravenousinfusion repeated daily. J Clin Oncol 2002;20:3772–84.

15. Herbst RS, Hess KR, Tran HT, et al. Phase I study of recombinanthuman endostatin in patients with advanced solid tumors. J Clin Oncol2002;20:3792–803.

16. Thomas JP, Arzoomanian RZ, Alberti D, et al. Phase I pharmacokineticand pharmacodynamic study of recombinant human endostatin in patientswith advanced solid tumors. J Clin Oncol 2003;21:223–31.

17. Hynes NE, Stern DF. The biology of erbB-2/neu/HER-2 and its role incancer. Biochim Biophys Acta 1994;1198:165–84.

18. Kraus MH, Issing W, Miki T, Popescu NC, Aaronson SA. Isolationand characterization of ERBB3, a third member of the ERBB/epidermalgrowth factor receptor family: evidence for overexpression in a subsetof human mammary tumors. Proc Natl Acad Sci U S A 1989;86:9193–7.

19. Plowman GD, Culouscou JM, Whitney GS, et al. Ligand-specificactivation of HER4/p180erbB4, a fourth member of the epidermal growthfactor receptor family. Proc Natl Acad Sci U S A 1993;90:1746–50.

20. Slamon DJ, Godolphin W, Jones LA, et al. Studies of the HER-2/neuproto-oncogene in human breast and ovarian cancer. Science 1989;244:707–12.

21. Penichet ML, Challita PM, Shin SU, Sampogna SL, Rosenblatt JD,Morrison SL. In vivo properties of three human HER2/neu -expressingmurine cell lines in immunocompetent mice. Lab Anim Sci 1999;49:179–88.

22. Kim YM, Jang JW, Lee OH, et al. Endostatin inhibits endothelial andtumor cellular invasion by blocking the activation and catalytic activity ofmatrix metalloproteinase. Cancer Res 2000;60:5410–3.

23. Shin SU, Wu D, Ramanathan R, Pardridge WM, Morrison SL.Functional and pharmacokinetic properties of antibody-avidin fusionproteins. J Immunol 1997;158:4797–804.

24. Challita-Eid PM, Penichet ML, Shin SU, et al. A B7.1-antibody fusionprotein retains antibody specificity and ability to activate via the T cellcostimulatory pathway. J Immunol 1998;160:3419–26.

25. Coloma MJ, Hastings A, Wims LA, Morrison SL. Novel vectors for theexpression of antibody molecules using variable regions generated bypolymerase chain reaction. J Immunol Methods 1992;152:89–104.

26. Shin SU, Morrison SL. Production and properties of chimeric antibodymolecules. Methods Enzymol 1989;178:459–76.

27. Vazquez F, Hastings G, Ortega MA, et al. METH-1, a human orthologof ADAMTS-1, and METH-2 are members of a new family of proteins withangio-inhibitory activity. J Biol Chem 1999;274:23349–57.

28. Iruela-Arispe ML, Lombardo M, Krutzsch HC, Lawler J, Roberts DD.Inhibition of angiogenesis by thrombospondin-1 is mediated by 2independent regions within the type 1 repeats. Circulation 1999;100:1423–31.

29. Pardridge WM, Boado RJ, Kang YS. Vector-mediated delivery of apolyamide(‘‘peptide’’) nucleic acid analogue through the blood-brainbarrier in vivo . Proc Natl Acad Sci U S A 1995;92:5592–6.

30. Kang YS, Pardridge WM. Use of neutral avidin improves pharmaco-

kinetics and brain delivery of biotin bound to an avidin-monoclonalantibody conjugate. J Pharmacol Exp Ther 1994;269:344–50.

31. Fenton BM, Paoni SF, Lee J, Koch CJ, Lord EM. Quantification oftumour vasculature and hypoxia by immunohistochemical staining andHbO2 saturation measurements. Br J Cancer 1999;79:464–71.

32. Gerber SA, Moran JP, Frelinger JG, Frelinger JA, Fenton BM, LordEM. Mechanism of IL-12 mediated alterations in tumour blood vesselmorphology: analysis using whole-tissue mounts. Br J Cancer 2003;88:1453–61.

33. Guo L, Burke P, Lo SH, Gandour-Edwards R, Lau D. Quantitativeanalysis of angiogenesis using confocal laser scanning microscopy.Angiogenesis 2001;4:187–91.

34. Karumanchi SA, Jha V, Ramchandran R, et al. Cell surface glypicansare low-affinity endostatin receptors. Mol Cell 2001;7:811–22.

35. Miosge N, Simniok T, Sprysch P, Herken R. The collagen type XVIIIendostatin domain is co-localized with perlecan in basement membranesin vivo. J Histochem Cytochem 2003;51:285–96.

36. Zorick TS, Mustacchi Z, Bando SY, et al. High serum endostatin levelsin Down syndrome: implications for improved treatment and prevention ofsolid tumours. Eur J Hum Genet 2001;9:811–4.

37. Augustin HG. Anti-angiogenic tumour therapy: will it work? TrendsPharmacol Sci 1998;19:216–22.

38. Folkman J. Anti-angiogenic gene therapy. Proc Natl Acad Sci U S A1998;95:9064–6.

39. Penichet ML, Kang YS, Pardridge WM, Morrison SL, Shin SU. Anantibody-avidin fusion protein specific for the transferrin receptor servesas a delivery vehicle for effective brain targeting: initial applications in anti-HIV antisense drug delivery to the brain. J Immunol 1999;163:4421–6.

40. Beck MT, Chen NY, Franek KJ, Chen WY. Prolactin antagonist-endostatin fusion protein as a targeted dual-functional therapeutic agentfor breast cancer. Cancer Res 2003;63:3598–604.

41. Pawliuk R, Bachelot T, Zurkiya O, Eriksson A, Cao Y, Leboulch P.Continuous intravascular secretion of endostatin in mice from transducedhematopoietic stem cells. Mol Ther 2002;5:345–51.

42. Eisterer W, Jiang X, Bachelot T, et al. Unfulfilled promise ofendostatin in a gene therapy-xenotransplant model of human acutelymphocytic leukemia. Mol Ther 2002;5:352–9.

43. Standker L, Schrader M, Kanse SM, Jurgens M, Forssmann WG,Preissner KT. Isolation and characterization of the circulating form ofhuman endostatin. FEBS Lett 1997;420:129–33.

44. Tanaka T, Cao YH, Folkman J. Viral vector-targeted antiangiogenicgene therapy utilizing an angiostatin complementary DNA. Cancer Res1998;58:3362–9.

45. Lin P, Buxton JA, Acheson A. Antiangiogenic gene therapy targetingthe endothelium-specific receptor tyrosine kinase Tie2. Proc Natl AcadSci U S A 1998;95:8829–34.

46. Griscelli F, Li H, Bennaceur-Griscelli A. Angiostatin gene transfer:inhibition of tumor growth in vivo by blockage of endothelial cellproliferation associated with a mitosis arrest. Proc Natl Acad Sci U S A1998;95:6367–72.

47. Kong HL, Crystal RG. Gene therapy strategies for tumor antiangio-genesis. J Natl Cancer Inst 1998;90:273–86.

48. Zabner J, Ramsey BW, Meeker DP. Repeat administration of anadenovirus vector encoding cystic fibrosis transmembrane conductanceregulator to the nasal epithelium of patients with cystic fibrosis. J ClinInvest 1996;97:1504–11.

49. Yei S, Mittereder N, Tang K. Adenovirus-mediated gene transfer forcystic fibrosis: quantitative evaluation of repeated in vivo vectoradministration to the lung. Gene Ther 1994;1:192–200.

50. Yang Y, Li Q, Ertl HC. Cellular and humoral immune responses to viralantigens create barriers to lung-directed gene therapy with recombinantadenoviruses. J Virol 1995;69:2004–15.

51. Sonneveld P, Pieters R. Immunophenotyping as a guide for targetedtherapy.Best Pract Res Clin Haematol 2003;16:629–44.

52. Pandit-Taskar N, Hamlin PA, Reyes S, Larson SM, Divgi CR. Newstrategies in radioimmunotherapy for lymphoma. Curr Oncol Rep 2003;5:364–71.

53. Cheson BD. Radioimmunotherapy of non-Hodgkin lymphomas. Blood2003;101:391–8.

54. Baselga J, Tripathy D, Mendelsohn J, et al. Phase II study of weekly





intravenous recombinant humanized anti-p185HER2 monoclonal antibodyin patients with HER2/neu -overexpressing metastatic breast cancer. J ClinOncol 1996;14:737–44.

55. Vogel CL, Cobleigh MA, Tripathy D, et al. Efficacy and safety oftrastuzumab as a single agent in first-line treatment of HER2-overexpressing metastatic breast cancer. J Clin Oncol 2002;20:719–26.

56. Burstein HJ, Harris LN, Marcom PK, et al. Trastuzumab andvinorelbine as first-line therapy for HER2-overexpressing metastatic breastcancer: multicenter phase II trial with clinical outcomes, analysis of serumtumor markers as predictive factors, and cardiac surveillance algorithm.J Clin Oncol 2003;21:2889–95.

57. Yakes FM, Chinratanalab W, Ritter CA, King W, Seelig S, Arteaga CL.Herceptin-induced inhibition of phosphatidylinositol-3 kinase and Akt isrequired for antibody-mediated effects on p27, cyclin D1, and antitumoraction. Cancer Res 2002;62:4132–41.

58. Liang K, Lu Y, Jin W, Ang KK, Milas L, Fan Z. Sensitization of breastcancer cells to radiation by trastuzumab. Mol Cancer Ther 2003;2:1113–20.

59. Sliwkowski MX, Lofgren JA, Lewis GD, Hotaling TE, Fendly BM, FoxJA. Nonclinical studies addressing the mechanism of action of trastuzu-mab (Herceptin). Semin Oncol 1999;26:60–70.

60. Pegram MD, Reese DM. Combined biological therapy of breast cancerusing monoclonal antibodies directed against HER2/neu protein andvascular endothelial growth factor. Semin Oncol 2002;29:29–37.

61. Viloria Petit AM, Rak J, Hung MC, et al. Neutralizing antibodiesagainst epidermal growth factor and erbB-2/neu receptor tyrosinekinases down-regulate vascular endothelial growth factor produc-tion by tumor cells in vitro and in vivo . Am J Pathol 1997;151:1523–151:1523–30.

62. Yang W, Klos K, Yang Y, Smith TL, Shi D, Yu D. ErbB2 over-expression correlates with increased expression of vascular endothelialgrowth factors A, C, and D in human breast carcinoma. Cancer 2002;94:2855–61.

63. Izumi Y, Xu L, di Tomaso E, Fukumura D, Jain RK. Herceptinacts as an anti-angiogenic cocktail. Nature 2002;416:279–80.

64. Clynes RA, Towers TL, Presta LG, Ravetch JV. Inhibitory Fc receptorsmodulate in vivo cytotoxicity against tumor targets. Nat Med 2000;6:443–6.

65. Kim KM, Shin EY, Moon JH, et al. Both the epitope specificity andisotype are important in the antitumor effect of monoclonal antibodiesagainst Her-2/neu antigen. Int J Cancer 2002;102:428–34.

66. Pegram MD, Konecny GE, O’Callaghan C, Beryt M, Pietras R,Slamon DJ. Rational combinations of trastuzumab with chemotherapeuticdrugs used in the treatment of breast cancer. J Natl Cancer Inst 2004;96:739–49.

67. Citrin D, Lee AK, Scott T, et al. In vivo tumor imaging inmice with near-infrared labeled endostatin. Mol Cancer Ther 2004;3:481–8.





Article on antibody-endostatinfusion protein

In the article on antibody-endostatin fusion protein in theMay 2005 issue (1), the last name of author Dr. HovavNechushtan was spelled incorrectly.

Reference

1. Cho H-M, Rosenblatt JD, Kang Y-S, et al. Enhance inhibition of murinetumor and human breast tumor xenografts using targeted delivery of anantibody-endostatin fusion protein. Mol Cancer Ther 2005;4:956–67.

Copyright C 2005 American Association for Cancer Research.doi:10.1158/1535-7163.MCT-correction

Correction

Mol Cancer Ther 2005;4(9). September 2005

1456

2005;4:956-967. Mol Cancer Ther Hyun-Mi Cho, Joseph D. Rosenblatt, Young-Sook Kang, et al. antibody-endostatin fusion proteintumor xenografts using targeted delivery of an Enhanced inhibition of murine tumor and human breast

Updated version

http://mct.aacrjournals.org/content/4/6/956

Access the most recent version of this article at:

Cited articles

http://mct.aacrjournals.org/content/4/6/956.full#ref-list-1

This article cites 65 articles, 31 of which you can access for free at:

Citing articles

http://mct.aacrjournals.org/content/4/6/956.full#related-urls

This article has been cited by 5 HighWire-hosted articles. Access the articles at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. (CCC)Click on "Request Permissions" which will take you to the Copyright Clearance Center's

.http://mct.aacrjournals.org/content/4/6/956To request permission to re-use all or part of this article, use this link



http://mct.aacrjournals.org/content/4/6/956.full#ref-list-1

http://mct.aacrjournals.org/content/4/6/956.full#related-urls

http://mct.aacrjournals.org/cgi/alerts

mailto:[email protected]



Enhanced inhibition of murine tumor and human breast tumor

Documents

Transcript of Enhanced inhibition of murine tumor and human breast tumor